Prior art has explored using bio-inspired models of human vision and other sensors for robot vision control. Other bio-inspired models of human vision have lead to robot control techniques that do not depend on classical negative feedback loops for control. Still separately, various methods have been proposed for translating an overall goal into specific robot control commands. These separate technologies have yet to be integrated into one overall method of robot control. There exists a need for control of a multimodal sensor robot responding to an overall plan or goal.